By marisaokano - 1/29/2014
I have encountered a few issues in running my experiment using inquisit. I have ran this experiment since the summer using the exact same scripts without any issues. I have not changed my scripts, only created a backup copy of the folder on my MAC, yet when I went to run my experiment after winter break (second semester university) inquisit has been failing.
- I have seven separate scripts all run together in 1 experiment
- the scripts include questionnaires, presentation of images, video stimuli, and an implicit association task
- the data has been collected for over 100 participants with no issues up until this point
- Inquisit is registered for use on a PC (lab version) computer (used by participants) this has NOT changed from last semester
- I switched my PERSONAL computer that I write/revise the scripts on over from a PC computer to a MAC last semester, however no issues arose from this switch prior to now.
- the issues arose before i updated my inquisit to the most recent version, I uploaded 4.0.4, and the issues are STILL occurring.
The issues I am encountering are:
- data is not recording
- individual scripts/data documents are slow to open (have to click multiple times/sometimes won't work)
- experiment will start, but freezes at random points in the experiment
- when the script freezes, the computer needs to be restarted
- the only data that will record is the summary data - no individual script data is recorded regardless of where the freeze occurs
- lastly - when trying to open each of my scripts I encounter an issue specific to my implicit task script - it says that the default language will not read the script and asks me to choose from a list of very confusing languages?
- again, I did NOT make any change to the scripts at all, so I am confused as to why this is occurring?
By Dave - 1/29/2014
#1: You should make sure you use the same, current Inquisit version on your development machine as well as on your data collection machines.
#2: You should make sure that your experiments are saved in *.iqx format, not the legacy *.exp format (open any existing *.exp and save in *.iqx format via File -> Save As...).
#3: Check the log files via Tools -> View Log File... for any apparent issues.
For anything more specific, you would have to make the respective scripts available.
By marisaokano - 1/30/2014
When you say to use the "same, current version", are you referring to the update alone (i.e. 4.0.4), or the version I use on my MAC (i.e. i downloaded the MAC version on my computer, but use the Desktop version on my experiment computer).
The files are saved in the correct format.
I have narrowed down the issues to be involved with two scripts:
- first: my implicit task script runs fine, but when I go to open the data - it won't open. The screen displays a command about not being able to read the language and to identify which language I originally encoded the script in (see below for pictures).
- second: my 4th script that runs in the batch is a presentation of video stimuli (the first 3 scripts run fine - only the 2nd script doesn't collect data (see issue one above)). The participant is shown a 5 second clip, then a survey page pops up and they answer questions about the video. 20 videos play in total. The issue is that the videos are now randomly freezing and the participant cannot proceed with the task. I can press control + B to skip the block, and a few more videos will play properly and then the frozen screen will happen again. So - in summary, as soon as a video freezes, the participant can no longer proceed with the task, and therefore do not complete the remaining scripts).
I have attached the implicit task script (pGNAT) and the associated data file. I also provided the video presentation script (TSV), however I cannot supply the videos as they contain confidential info. The videos are in wmv format and range between 3.2-4.4 MB. I have run these exact same videos with no problems prior to this semester.
By Dave - 1/30/2014
#1: You should use the same version -- right now: 126.96.36.199 -- both on your Mac as well as on the data collection machines. The version number (188.8.131.52) is identical for both Mac and Windows (cf. https://www.millisecond.com/download/).
#2: Inquisit data files are merely tab-delimited plain text files (you can simply take a given *.iqdat file, change the extension to *.txt and view it in any text editor). I have attached an example file you can use to try that out and see for yourself.
I'm not sure what happened with or to that GNAT *.iqdat file you attached. It certainly does not look like a typical Inquisit data file and appears to essentially contain binary data. Did you at some point during the data collection period set /encrypt=true in the GNAT script's <data> element and remove the /encrypt attribute at a later point?
FWIW, I have never seen anything like this happen to a data file and frankly have no idea (save for the /encrypt hypothesis mentioned above) what could have caused this. If you have any additional relevant information, please share!
In any case, to have the script record data (again), it should be sufficient to remove or rename the existing "2H. pGNAT.iqdat" in the script's folder to get a fresh file. Alternatively, you can have any script output separate data files per participant via the <data> element's /separatefiles attribute.
#3: It is unfortunately very hard (if not impossible) to diagnose the video freezing issue without access to the actual video files. (a) Could you please confirm that your data collection machines' Inquisit installations are up to date (i.e. at version 184.108.40.206), and if they are not, please update them? (b) Could you please provide the technical log file accessible via Tools -> View Log File... from one of the machines affected?
By marisaokano - 2/3/2014
I have ensured that both computers are using the 220.127.116.11 version. I changed the original pGNAT .iqdat file to .txt, however the data was still illegible (i.e. binary and symbols). Perhaps I misunderstood your instructions? When you open the pGNAT data file, does your computer ask you the original language the script was encoded in? I originally wrote the script on a PC laptop using the 4.0.3 version of Inquisit, is there a default language that this would have been encoded in, or is there a way for me to determine the original language of encoding?
As for the encrypt attribute, unfortunately I have not used this in any of my script at any point in the development, so that couldn't be what contributed to this situation. I did create a fresh iqdat file for the gNAT and it is collecting data properly now, so thank you for that tip! Now it is just a matter of me being able to use the data from last semester.
For the videos, I have narrowed it down to three of the 20 that freeze. The videos are presented at random, therefore I do not know which exact videos they are (I can determine that by removing the random assignment, but I won't do that unless necessary); however I have not changed anything about the videos since last semester so I am unsure as to why they are freezing now. A few videos will play properly, and then when the video that freezes is presented, it doesn't go past a black box in the middle of the screen (which is the same size and position of the video that is presented). I have attached the technical file log for you to look at.
By Dave - 2/3/2014
Yes, I'm getting the same prompt as you for your corrupted data file. That's because, for all intents and purposes, it isn't an Inquisit data file and the application doesn't know what to do with it. I don't know what happened to that file.
The instructions re. changing a data files extension to *.txt was merely meant to illustrate to you how a proper data file should look and that there's nothing proprietary about its format -- they're just simple text files, which can be read by virtually any application, not just Inquisit.
Thanks for posting the log file re. the freezing videos. The log shows a bunch of unknown DirectShow execution errors, which are probably related to the problem at hand:
2014-02-03T11:01:28.493 InquisitLab 18.104.22.168 2311 Script='E:/Marisa/FoA Honours/4H. TSV.iqx' GroupID=1 SubjectID=TEST ExecutionError Inquisit Media Error: Unknown DirectShow error. line 112: file win\MediaPlayerControl.cpp
They don't provide much meaningful insight, though, since they are "unknown". As you say the freezing appears to be limited to 3 particular files in your set, can you tell me if there is something different about those files compared to the other ones in your set which are working fine?
By marisaokano - 2/5/2014
That is so odd about that pGNAT data file.... I will keep investigating into possible reasons why that happened and if I find anymore information I will let you know. For now it looks as though I will just have to classify that data as lost due to error. Out of the seven scripts, the pGNAT was the only one to have the language issue. Thank you for further explaining the .txt file, it is helpful to know that there is nothing proprietary about the format.
As for the video files, I have taken a backup version of the video clips and moved them into a separate folder with the TSV script. All videos played fine on my MAC (I attached the log). To test the PC desktop computer, I am going to replace the video files on the USB I use to run the experiment with these backup clips and see whether freezing still occurs. I have not identified any differences between the three videos that freeze and the rest, therefore I am just going to try and replace all videos. If this does not work, I may convert the video files into a different format and test the script.